Number Of True Negatives Research Articles

Integrating old and new complexity measures toward automated seizure detection from long-term video EEG recordings

SummaryAutomated seizure detection in long-term video-EEG recordings is far from being integrated into common clinical practice. Here, we leverage classical and state-of-the-art complexity measures to robustly and automatically detect seizures from scalp recordings. Brain activity is scored through eight features, encompassing traditional time domain and novel measures of recurrence. A binary classification algorithm tailored to treat unbalanced dataset is used to determine whether a time window is ictal or non-ictal from its features. The application of the algorithm on a cohort of ten adult patients with focal refractory epilepsy indicates sensitivity, specificity, and accuracy of 90%, along with a true alarm rate of 95% and less than four false alarms per day. The proposed approach emphasizes ictal patterns against noisy background without the need of data preprocessing. Finally, we benchmark our approach against previous studies on two publicly available datasets, demonstrating the good performance of our algorithm.

Short communication: Association of milk fatty acids with early lactation hyperketonemia and elevated concentration of nonesterified fatty acids

The objective of our study was to extend the limited research available on the association between concentrations of milk fatty acids and elevated nonesterified fatty acids (NEFA) and β-hydroxybutyrate (BHB) concentrations in early lactation dairy cattle. Measurement of milk fatty acids for detection of cows in excessive negative energy balance has the potential to be incorporated in routine in-line monitoring systems. Blood samples were taken from 84 cows in second or greater lactation 3 times per week between 3 to 14 d in milk. Cows were characterized as hyperketonemic (HYK) if blood BHB concentration was ≥1.2mmol/L at least once and characterized as having elevated concentrations of NEFA (NEFAH) if serum NEFA concentration was ≥1mmol/L at least once. Composition of colostrum and milk fatty acids at wk 2 postpartum was used to investigate the potential diagnostic value of individual fatty acids and fatty acid ratios for the correct classification of cows with NEFA and BHB concentrations above these thresholds, respectively. Receiver operating characteristic (ROC) curves were used to identify thresholds of fatty acid concentration and fatty acid ratios when ROC area under the curve was ≥0.70. Correct classification rate (CCR, %) was calculated as {[(number of true positives + number of true negatives)/total number tested] × 100}. None of the colostrum fatty acids yielded a sufficiently high area under the curve in ROC analysis for the association with HYK and NEFAH. The following fatty acids and fatty acid ratios were identified for an association with NEFAH (threshold, CCR): C15:0 (≤0.65g/100g, 68.3%); cis-9 C16:1 (≥1.85g/100g, 70.7%); cis-9 C18:1 (≥26g/100g, 69.5%), cis-9 C18:1 to C15:0 ratio (≥45, 69.5%); cis-9 C16:1 to C15:0 (≥2.50, 73.2%). Several fatty acids were associated with HYK (threshold, CCR): C6:0 (≤1.68g/100g, 80.5%), C8:0 (≤0.80g/100g, 80.5%), C10:0 (≤1.6g/100g, 79.3%); C12:0 (≤1.42g/100g, 82.9%); C14:0 (≤6.10g/100g, 84.1%); C15:0 (≤0.50g/100g, 82.9%), cis-9 C18:1 (≥30g/100g, 81.7%). The use of fatty acid ratios did not improve CCR over using individual fatty acids for the classification of HYK. Colostrum fatty acid composition was not useful in predicting NEFAH or HYK between 3 to 14 d in milk. Accuracy of milk fatty acids and fatty acid ratios to correctly classify cows with elevated concentrations of NEFA and BHB between 3 to 14 d in milk was moderate and overall higher for HYK. Determining changes in the fatty acid composition of milk fat from milk samples at wk 2 postpartum for the detection of cows with elevated concentrations of BHB and NEFA can currently not be recommended to replace direct measurement. Future applications should target repeated milk sampling between 3 to 14 d in milk to identify the best sampling for determination of milk fatty acid composition within the first 2 wk postpartum.

Expressions of matrix metalloproteinase-9 (MMP-9), dentin sialophosphoprotein (DSPP), and osteopontin (OPN) at histologically negative surgical margins may predict recurrence of oral squamous cell carcinoma

Up to 50% of oral squamous cell carcinomas (OSCCs) recur following surgical resections with conventional "histologically-negative" margins. Three members of the SIBLING family of proteins: dentin sialophophoprotein (DSPP); bone sialoprotein (BSP); and osteopontin OPN are upregulated in OSCCs. In this study, we aimed to correlate the expression of DSPP, OPN and BSP as well as three SIBLING-partners, matrix metalloproteinase-2 (MMP-2), matrix metalloproteinase-3 (MMP-3), and matrix metalloproteinase-9 (MMP-9), at histologically-negative margins of OSCCs with tumor recurrence. Immunohistochemical analyses of the SIBLINGs and MMP expressions at histologically-negative margins of OSCC was carried out in a retrospective study of 20 patients, and the results correlated with tumor recurrence. Each protein was dichotomized as "present" (≥10% staining) or "absent" (more than 10% staining). The Sensitivity, Specificity, Positive Predictive Value(PV+) and Negative Predictive Value (PV-) for recurrence was calculated for each protein, along with their overall diagnostic accuracy, calculated as: (number of true positives + number of true negatives)/ number of patients. OSCC recurred in 9 of 20 patients (45%), a ratio not significantly different from the estimated population recurrence rate of 50% (p = 0.664). Among the SIBLINGs, DSPP and OPN showed the greatest Accuracy with DSPP being more Sensitive (89%) and OPN more Specific (64%). MMP-9 showed the greatest overall Accuracy (80%), slightly less Sensitivity (67%) and more Specificity (100%), than either DSPP or OPN. MMP-9 showed a superior positive PV than either DSPP or OPN. The negative PVs of OPN and MMP-9 were almost identical, and inferior to DSPP. We conclude that DSPP, OPN, or MMP-9 expressions at histologically-negative surgical margins predict OSCC recurrence with MMP-9 being the preferred predictor. These proteins may identify patients who could benefit from more extensive resection, or from adjunct treatments such as radiotherapy.

High test–retest reliability of checkerboard reversal visual evoked potentials (VEP) over 8 months

The inter-individual variation in the shape of a visual evoked potential (VEP) is large even for simple stimuli. We compared the inter-individual variation in VEP waveform to the intra-individual stability. We recorded VEP with checkerboard stimulation in 10 women aged 19-29years in two sessions. We determined the latencies and the peak-to-peak amplitude of N75 and P100. As a new approach in VEP research, we regressed VEP waveforms pairwise onto each other and calculated a t-value between all sessions. The maximal t-value was taken to indicate recognition for all 19 comparisons performed with one session. The recognition rate was cross-validated in a generalized linear regression model (GLM). The number of sessions correctly matched to the correct subject was 19 of 20 (true positives) leading to a sensitivity of 95.0% with confidence interval [75.1% 99.9%] for the method. The number of true negatives was 359 of 360 leading to a specificity of 99.7% [98.5% 100.0%]. The VEP waveform shows high intra-individual stability compared to the inter-individual variation in healthy women. With a new statistical approach the effect of external factors on the VEP waveform can now be contrasted against the normal variability over time in longitudinal studies.

Near-Infrared Laser Spectroscopy as a Screening Tool for Detecting Hematoma in Patients with Head Trauma

Among imaging techniques, computed tomography (CT) is a reliable method for detecting intracranial hematomas in patients with head trauma, but it is not generally available in special circumstances like prehospital situations and harsh conditions such as those following an earthquake. The objective of this study is to determine if near-infrared spectroscopy (NIRS) is useful for performing CT scans on patients with closed head trauma that present to medical centers that do not have the ability to perform a CT scan or in prehospital or harsh situations. Near-infrared spectroscopy and CT scan were compared according to sensitivity and specificity. This was an observational, prospective study. One hundred forty-eight patients admitted to Rasool Akram General Hospital in Tehran, Iran with head injuries during a one-month period were studied using NIRS and CT. The observational, prospective study was conducted and sensitivity, specificity, positive and negative predictive values of NIRS were calculated. Chi-square and Kappa analysis was performed, and a p-value <0.05 was considered significant. According to the CT scan findings, 54 (36.5%) of the patients developed intracranial hematoma. The NIRS examination showed that 69 patients (46.6%) might have intracranial hematoma. The number of true negatives was 73 and the number of false negatives was six patients. The sensitivity and specificity of NIRS examination was 88.9% and 77.7%, respectively. This study speculates that NIRS may be a useful screening tool to detect intracranial hematoma. This capability could be useful in special situations like in a deprived area, medical centers without CT scan capabilities, prehospital situations, and in harsh conditions like those after an earthquake or other disasters that causes increased numbers of victims with closed head trauma.

Highly Efficient Classification and Identification of Human Pathogenic Bacteria by MALDI-TOF MS

Accurate and rapid identification of pathogenic microorganisms is of critical importance in disease treatment and public health. Conventional work flows are time-consuming, and procedures are multifaceted. MS can be an alternative but is limited by low efficiency for amino acid sequencing as well as low reproducibility for spectrum fingerprinting. We systematically analyzed the feasibility of applying MS for rapid and accurate bacterial identification. Directly applying bacterial colonies without further protein extraction to MALDI-TOF MS analysis revealed rich peak contents and high reproducibility. The MS spectra derived from 57 isolates comprising six human pathogenic bacterial species were analyzed using both unsupervised hierarchical clustering and supervised model construction via the Genetic Algorithm. Hierarchical clustering analysis categorized the spectra into six groups precisely corresponding to the six bacterial species. Precise classification was also maintained in an independently prepared set of bacteria even when the numbers of m/z values were reduced to six. In parallel, classification models were constructed via Genetic Algorithm analysis. A model containing 18 m/z values accurately classified independently prepared bacteria and identified those species originally not used for model construction. Moreover bacteria fewer than 10(4) cells and different species in bacterial mixtures were identified using the classification model approach. In conclusion, the application of MALDI-TOF MS in combination with a suitable model construction provides a highly accurate method for bacterial classification and identification. The approach can identify bacteria with low abundance even in mixed flora, suggesting that a rapid and accurate bacterial identification using MS techniques even before culture can be attained in the near future.

The diagnostic value of PET/CT scanning in patients with cervical cancer: A prospective study

Objective To investigate the clinical value of PET/CT as a supplement to FIGO staging in patients with cervical cancer stage ≥ 1B. Methods This prospective study included 120 consecutive patients. After staging, a whole-body PET/CT scan was performed and these examinations were divided into two groups: (1) patients suitable for radical hysterectomy including lymph node dissection and (2) patients referred to combined chemo/radiation therapy. The results were compared to histopathological findings and/ or follow-up. Results Twenty-seven patients underwent radical surgery; four of these had PET/CT scans revealing pathological foci in the pelvis. Three (11%) were true positive; one was false positive. Twenty-two patients had true negative PET/CT scans concerning pelvic lymph nodes. One patient had a false negative node. For these patients, we found the positive predictive value (PPV) to be 75%, negative predictive value (NPV) 96%, sensitivity 75%, specificity 96%. Regarding para-aortal nodal disease in the total population of 119 patients, 15 patients had true positive scans. The number of true negatives was 103, resulting in PPV 94%, NPV 100%, sensitivity 100% and specificity 99%. PET/CT scans showed distant metastases in 19 patients, 10 were true positive and nine were false positive. The remaining 100 patients were considered true negative for distant metastases and for these patients, we found PPV 63%, NPV 100%, sensitivity 100% and specificity 94%. Conclusions Whole-body FDG PET/CT scanning for newly diagnosed cervical cancer FIGO stage ≥ 1B has a high sensitivity and specificity, and can be a valuable supplement to the FIGO staging procedure.

Estimators of sensitivity and specificity in the presence of verification bias: A Bayesian approach

Verification bias can occur if some of the patients with test results are not selected to receive the gold standard procedure. Unverified cases frequently are not suggestive to be positives. Consequently, the set of verified cases overestimates the number of true positives and underestimates the number of true negatives. The sensitivity and specificity estimates based only on the patients with verified disease are often biased. In this article we derive estimators for sensitivity and specificity not subject to verification bias using a Bayesian approach. Marginal posterior densities of all parameters are estimated using the Gibbs sampler algorithm. An application to the study of accuracy of Hybrid Capture II in the diagnosis of cervical intraepithelial neoplasia grades 2 and 3 illustrates the proposed methodology.

The Mini Mental State Exam as a predictor of neuropsychological functioning after cardiac surgery.

The present longitudinal study was designed to: 1) determine the ability of the Mini Mental State Exam (MMSE) to predict neuropsychologic impairment based on neuropsychologic testing five to seven days and six weeks after cardiac surgery; and 2) to determine whether the traditional or the education-related MMSE norms are more appropriate to use for this purpose. The day before surgery (T1), before hospital discharge (T2), and six weeks after surgery (T3), 247 subjects completed a battery of five neuropsychologic tests. Subjects also completed the Center for Epidemiological Studies Depression Scale and the Speilberger State-Trait Anxiety Inventory. Subjects completed the MMSE two to three days after surgery. Stepwise regression analyses revealed that the MMSE significantly predicted only a small portion of the variance in neuropsychologic test performance at T2, and to an even lesser extent at T3, over and above the demographic variables. In assessing the association between an impairment score (derived from the neuropsychologic test battery) and the MMSE, we found that the traditional MMSE cut-off score maximized specificity (number of true negatives) while the education-adjusted MMSE cut-off scores maximized sensitivity (number of true positives). These results suggest that although the MMSE is widely used to assess cognitive mental status, it may have limited value in identifying patients with cognitive impairment post-cardiac surgery, and special attention must be paid to the cut-off scores used in interpreting the MMSE.